Diabetes mellitus is a serious and chronic metabolic disease that is characterized by high blood glucose (hyperglycemia) and affects millions of people world-wide. SGLT2 is a Sodium-dependent GLucose co-Transporter protein which affects the reabsorption of glucose in the kidney. It is estimated that 90% of renal glucose reabsorption is facilitated by SGLT2. Since glucose reabsorption is mediated predominantly by SGLT2 and because high glucose levels have been identified as a cause of disease in diabetes, SGLT2 has become a drug target for type 2 diabetes therapy. Selective inhibition of SGLT2 has the potential to reduce hyperglycemia by inhibiting glucose reabsorption in the kidney with elimination of glucose by excretion in the urine (glucosuria).
A significant number of SGLT2 inhibitors are currently in clinical development and a significant portion of these are β-C-arylglucosides. Of these, dapagliflozin, developed by Bristol-Myers Squibb and AstraZeneca, is at the most advanced stage of the development with its new drug application (NDA) having been accepted for review by the Food and Drug Administration (FDA) in 2011.
In addition to dapagliflozin, there are a significant number of other β-C-arylglucoside derived drug candidates, most of which differ only in the aglycone moiety (i.e., these compounds comprise a central 1-deoxy-glucose ring moiety that is arylated at C1). It is this fact that makes them attractive targets for a novel synthetic platform technology, since a single methodology should be able to furnish a plurality of products. Among β-C-arylglucosides that possess known SGLT2 inhibition also currently in clinical development are canagliflozin, empagliflozin, and ipragliflozin.

A series of synthetic methods have been reported in the peer-reviewed and patent literature that can be used for the preparation of β-C-arylglucosides. These methods are described below and are referred herein as the gluconolactone method, the metalated glucal method, the glucal epoxide method and the glycosyl leaving group substitution method.
The gluconolactone method: In 1988 and 1989 a general method was reported to prepare C-arylglucosides from tetra-O-benzyl protected gluconolactone, which is an oxidized derivative of glucose (see J. Org. Chem. 1988, 53, 752-753 and J. Org. Chem. 1989, 54, 610-612). The method comprises: 1) addition of an aryllithium derivative to the hydroxy-protected gluconolactone to form a hemiketal (a.k.a., a lactol), and 2) reduction of the resultant hemiketal with triethylsilane in the presence of boron trifluoride etherate. Disadvantages of this classical, but very commonly applied method for β-C-arylglucoside synthesis include:                1) poor “redox economy” (see J. Am. Chem. Soc. 2008, 130, 17938-17954 and Anderson, N. G. Practical Process Research & Development, 1st Ed.; Academic Press, 2000 (ISBN-10: 0120594757); pg 38)—that is, the oxidation state of the carbon atom at C1, with respect to glucose, is oxidized in the gluconolactone and then following the arylation step is reduced to provide the requisite oxidation state of the final product.        2) due to a lack of stereospecificity, the desired β-C-arylglucoside is formed along with the undesired α-C-arylglucoside stereoisomer (this has been partially addressed by the use of hindered trialkylsilane reducing agents (see Tetrahedron: Asymmetry 2003, 14, 3243-3247) or by conversion of the hemiketal to a methyl ketal prior to reduction (see J. Org. Chem. 2007, 72, 9746-9749 and U.S. Pat. No. 7,375,213)).        

The metalated glucal method: U.S. Pat. No. 7,847,074 discloses preparation of SGLT2 inhibitors that involves the coupling of a hydroxy-protected glucal that is metalated at C1 with an aryl halide in the presence of a transition metal catalyst. Following the coupling step, the requisite formal addition of water to the C-arylglucal double bond to provide the desired C-aryl glucoside is effected using i) hydroboration and oxidation, or ii) epoxidation and reduction, or iii) dihydroxylation and reduction. In each case, the metalated glucal method represents poor redox economy because oxidation and reduction reactions must be conducted to establish the requisite oxidation states of the individual C1 and C2 carbon atoms.
U.S. Pat. Appl. 2005/0233988 discloses the utilization of a Suzuki reaction between a C1-boronic acid or boronic ester substituted hydroxy-protected glucal and an aryl halide in the presence of a palladium catalyst. The resulting 1-C-arylglucal is then formally hydrated to provide the desired 1-C-arylglucoside skeleton by use of a reduction step followed by an oxidation step. The synthesis of the boronic acid and its subsequent Suzuki reaction, reduction and oxidation, together, comprise a relatively long synthetic approach to C-arylglucosides and exhibits poor redox economy. Moreover, the coupling catalyst comprises palladium which is toxic and therefore should be controlled to very low levels in the drug substance.

The glucal epoxide method: U.S. Pat. No. 7,847,074 discloses a method that utilizes an organometallic (derived from the requisite aglycone moiety) addition to an electrophilic epoxide located at C1-C2 of a hydroxy-protected glucose ring to furnish intermediates useful for SGLT2 inhibitor synthesis. The epoxide intermediate is prepared by the oxidation of a hydroxy-protected glucal and is not particularly stable. In Tetrahedron 2002, 58, 1997-2009 it was taught that organometallic additions to a tri-O-benzyl protected glucal-derived epoxide can provide either the α-C-arylglucoside, mixtures of the α- and β-C-arylglucoside or the β-C-arylglucoside by selection of the appropriate counterion of the carbanionic aryl nucleophile (i.e., the organometallic reagent). For example, carbanionic aryl groups countered with copper (i.e., cuprate reagents) or zinc (i.e., organozinc reagents) ions provide the β-C-arylglucoside, magnesium ions provide the α- and β-C-arylglucosides, and aluminum (i.e., organoaluminum reagents) ions provide the α-C-arylglucoside.

The glycosyl leaving group substitution method: U.S. Pat. No. 7,847,074, also disclosed a method comprising the substitution of a leaving group located at C1 of a hydroxy-protected glucosyl species, such as a glycosyl halide, with a metalated aryl compound to prepare SGLT2 inhibitors. U.S. Pat. Appl. 2011/0087017 disclosed a similar method to prepare the SGLT2 inhibitor canagliflozin and preferably diarylzinc complexes are used as nucleophiles along with tetra-O-pivaloyl protected glucosylbromide.

Methodology for alkynylation of 1,6-anhydroglycosides reported in Helv. Chim. Acta. 1995, 78, 242-264 describes the preparation of 1,4-dideoxy-1,4-diethynyl-β-D-glucopyranoses (a.k.a., glucopyranosyl acetylenes), that are useful for preparing but-1,3-diyne-1,4-diyl linked polysaccharides, by the ethynylating opening (alkynylation) of partially protected 4-deoxy-4-C-ethynyl-1,6-anhydroglucopyranoses. The synthesis of β-C-arylglucosides, such as could be useful as precursors for SLGT2 inhibitors, was not disclosed. The ethynylation reaction was reported to proceed with retention of configuration at the anomeric center and was rationalized (see Helv. Chim. Acta 2002, 85, 2235-2257) by the C3-hydroxyl of the 1,6-anhydroglucopyranose being deprotonated to form a C3-O-aluminium species, that coordinated with the C6-oxygen allowing delivery of the ethyne group to the β-face of the an oxycarbenium cation derivative of the glucopyranose. Three molar equivalents of the ethynylaluminium reagent was used per 1 molar equivalent of the 1,6-anhydroglucopyranose. The ethynylaluminium reagent was prepared by the reaction of equimolar (i.e., 1:1) amounts of aluminum chloride and an ethynyllithium reagent that itself was formed by the reaction of an acetylene compound with butyllithium. This retentive ethynylating opening method was also applied (see Helv. Chim. Acta. 1998, 81, 2157-2189) to 2,4-di-O-triethylsilyl-1,6-anhydroglucopyranose to provide 1-deoxy-1-C-ethynyl-β-D-glucopyranose. In this example, 4 molar equivalents of the ethynylaluminium reagent was used per 1 molar equivalent of the 1,6-anhydroglucopyranose. The ethynylaluminium regent was prepared by the reaction of equimolar (i.e., 1:1) amounts of aluminum chloride and an ethynyl lithium reagent that itself was formed by reaction of an acetylene compound with butyllithium.
It can be seen from the peer-reviewed and patent literature that the conventional methods that can be used to provide C-arylglucosides possess several disadvantages. These include (1) a lack of stereoselectivity during formation of the desired anomer of the C-arylglucoside, (2) poor redox economy due to oxidation and reduction reaction steps being required to change the oxidation state of C1 or of C1 and C2 of the carbohydrate moiety, (3) some relatively long synthetic routes, (4) the use of toxic metals such as palladium, and/or (5) atom uneconomic protection of four free hydroxyl groups. With regard to the issue of redox economy, superfluous oxidation and reduction reactions that are inherently required to allow introduction of the aryl group into the carbohydrate moiety of the previously mention synthetic methods and the subsequent synthetic steps to establish the required oxidation state, besides adding synthetic steps to the process, are particular undesirable for manufacturing processes because reductants can be difficult and dangerous to operate on large scales due to their flammability or ability to produce flammable hydrogen gas during the reaction or during work-up, and because oxidants are often corrosive and require specialized handling operations (see Anderson, N. G. Practical Process Research & Development, 1st Ed.; Academic Press, 2000 (ISBN-10: 0120594757); pg 38 for discussions on this issue).
In view of the above, there remains a need for a shorter, more efficient and stereoselective, redox economic process for the preparation of β-C-arylglucosides. A new process should be applicable to the industrial manufacture of SGLT2 inhibitors and their prodrugs, or for other medicinally useful drug candidates or drugs, or for synthetic building blocks for non-medicinal uses. This invention addresses those needs.